This paper examines the feasibility of using commercial out-of-the-box Reconfigurable Field Programmable Gate Array (FPGA) technology and the OpenCL framework to create efficient Sobel edge-detection implementation, which is considered a fundamental part in the field of image and video processing. The revised proposed approach was created at a high level of abstraction and executed on high commodity Intel FPGA platform. This was performed in a manner that was designed to allow the high-level compiler/synthesis tool to manipulate a task a parallelism model. The most promising FPGA and the naive implementations were compared to their single-core CPU software equivalents while manipulating local-memory, pipelining, loop unrolling, vectorization, internal channels mechanisms and memory coalescing to provide a much more effective hardware design. The run-time and the power consumption attributes were estimated for each implementation. The proposed FPGA based implementations were found to have significantly better runtime and power consumption with approximately up to 37 folds of improvement in the whole execution/transfer time, and up to 53 folds of improvement in energy consumption when compared to a specific single-core CPU based implementation.
This article introduces a current research and development project in the field of Human Language Technologies which focuses on five of the oficial languages of South Afi-ica. The basic components of an automatic information retrieval system are explicated with special reference to the types of speech databases that need to be developed in the South Afiican context. The position of Black South Afiican English (BSAE) is reviewed and an experiment is described in which the automatic speech recognition rate of two versions of BSAE were assessed, utilising state-ofthe-art technology. It is argued that the acquisition of BSAE speech databases is not only imperative for technology development, but that it may also make a contribution towarch the debate on the possible existence offirst language-bound varieties of BSAE. Hierdie artikel stel 'n lopende navorsings-en ontwikkelingsprojek in die veld van Menslike Taaltegnologie bekend, waarin daar op vyf van die amptelike tale van Suid-Afiika gefokus word. Die basiese komponente van 'n outomatiese informasieherwinningsisteem word uiteengesit met besondere verwysing na die tipes spraakdatabasisse wat in die SuidAfiikaanse konteks ontwikkel moet word. Die posisie van Swart SuidAfi-ikaanse Engels (SSAE) word bespreek en 'n eksperiment word beskryf waarin die outomatiese spraakherkenningsukses van twee vorme van SSAE bepaal word deur van intydse tegnologie gebruik te maak. Daar word geargumenteer dat die verkryging van SSAE-spraakdatabasisse nie net essensieel is vir tegnologieontwikkeling nie, maar oak 'n bydrae kan laver tot die debat oor die moontlike bestaan van vorme van SSAE wat aan verskillende eerstetale gebonde is. ;~w w w .~
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.